在本論文中,我們使用CIELAB標準色彩空間取代傳統的RGB色彩空間來進行多彩圖像辨識。LAB中的三分量分別代表亮度,紅、綠及黃、藍互補色。其中,本篇論文是使用多通道的聯合轉換相關器作為光學辨識的架構。而為了達到辨識程序中的形變不變特性,我們亦利用了最小平均相關能量法來尖銳化輸出峰值。此外,由於設備的成本考量,我們加入了影像編碼技術並比較使用後的結果。 從數據化結果可知,我們進行LAB、RGB與HSV在不同通道數目下,即三通道與兩通道的辨識。其後,我們加入了影像編碼技術並觀察此技術對辨識能力的影響。我們發現CIELAB空間的辨識效果普遍上都優於RGB色彩空間,與HSV的比較則是視情況而定。所以可得知在LAB表色系統模型下的彩色影像辨識能力是可被接受的。
In this thesis, the CIELAB standard color vision model instead of the traditional RGB color model is utilized for polychromatic pattern recognition. The L, A and B represents the lightness, the color red-green and yellow-blue, respectively. Here, the multi-channel joint transform correlator is set to be the optical discrimination configuration. To achieve the distortion invariance in discrimination processes, we also use the minimum average correlation energy approach to yield sharp correlation peak. Besides, the image encoding technique is introduced and compared because of the cost of the device. From the numerical results, we perform the recognition compared with HSV and RGB in different channel amounts, i.e. three and two selected channels. Subsequently, the encoding technique is adopted to observe the effects on discrimination quality. We discover that the recognition results based on CIELAB model are superior to RGB generally, and case by case with HSV. So we realize that the recognition ability based on CIELAB color specification system is accepted.